Disturbance attenuation control of active suspension with non-linear actuator dynamics

Sliding-mode output feedback control for active suspension with nonlinear actuator dynamics

“…Usually, active force f a is generated by a hydraulic actuator placed between the sprung and unsprung masses, therefore, a hydraulic actuator of a four-way valve-piston system is regarded. Detailed introduction of such a hydraulic actuator can be found in Lin and Kanellakopoulos (1997), Fialho and Balas (2002), and Ma and Chen (2011). The following basic concepts are adapted from them.…”

“…The control problem of active suspensions has attracted many researchers' attention as a result (see Yagiz and Hacioglu, 2008;Gao et al, 2010;Ma and Chen, 2011;Yao et al, 2013 and references therein). It is important for active suspensions to minimize the vertical forces transmitted to the drivers (i.e.…”

“…Maciejewski (2012) proposed a control method for a single-degree-of-freedom seat with a simple pneumatic suspension. Ma and Chen (2011) developed a disturbance attenuation control of active suspension with nonlinear actuator dynamics, 1 Jiangsu Province Key Laboratory of Aerospace Power Systems, Nanjing University of Aeronautics and Astronautics, China 2 College of Engineering, Nanjing Agricultural University, China but the authors did not discuss the effect of uncertainties, which probably happens. For active suspension systems with hydraulic actuators, because of their strong nonlinear property, sliding-mode control (SMC), which is effective for nonlinear and parameter-uncertain systems (Utkin, 1977;Slotine and Li, 1991;Hung et al, 1993;Gao, 1998), has attracted the interest of many researchers recently (Choi and Han, 2007;Benelghali et al, 2011;Lian, 2013).…”

Distributed mixed continuous‐discrete receding horizon filter for multisensory uncertain active suspension systems with measurement delays

“…Owing to the changing of the condition of road roughness, tolerance of the equipment, and measurement faults, the damping and spring coefficients (c, k 1 and k 2 ) can contain uncertainty. They are Table 1 Active suspension model parameters, noise statistics and initial conditions [7] Parameters Value mass of vehicle body m 1 , kg 1800 mass of vehicle wheel m 2 , kg 54 system noise intensities q 1 , q 2 220 measurement noise intensities r (i) 1 initial conditionsx 0 = [1 0.5] T , P 0 = diag [5,3] In this example, we focus on measuring the vertical displacements of the vehicle body and a wheel as a target object to implement the discussed FRHF and Lee's filter referred to as the DFF [7] on the three subperiods. Comparison results between true vertical displacements of the vehicle body x 1 (t) and wheel x 2 (t) and their estimates for the common receding horizon length com = i = 5 and measurement sampling time interval t = t k − t k−1 = 0.1 are shown in Figs.…”

“…In recent years, a lot of solutions have been proposed to improve the suspension performance [2][3][4][5][6][7]. Among them, Kalman filtering algorithm for vehicle active suspension system with time-delays is proposed by Lee et al [7].…”